DsPy crash course - optimizing LLM pipelines with DsPy (part 2)
The main idea of DsPy is to optimize not just your LLM model but the end-to-end pipeline. This video walks through a simple implementation of a multi-hop question-answering pipeline using DsPy. More specifically, we are going to implement the Baleen pipeline and optimize it using the BootStrap Few Shot algorithm by compiling the DsPy program.
Hope it's useful!
RELATED LINKS
Implemented Notebook - https://github.com/ai-bites/generativ...
Baleen paper - https://arxiv.org/abs/2101.00436
DsPy Docs - https://dspy-docs.vercel.app/docs/intro
⌚️ ⌚️ ⌚️ TIMESTAMPS ⌚️ ⌚️ ⌚️
0:00 - Intro
1:00 - Architecture of Baleen
1:40 - Baleen Implementation Overview
2:32 - Phi LM and Colbert V2 Retrieval Model
3:02 - HotPotQA Dataset
6:02 - Evaluation Metric
7:12 - Compiling the DsPy Pipeline
MY KEY LINKS
YouTube: / @aibites
Twitter: / ai_bites
Patreon: / ai_bites
Github: https://github.com/ai-bites
WHO AM I?
I am a Machine Learning researcher/practitioner who has seen the grind of academia and start-ups. I started my career as a software engineer 15 years ago. Because of my love for Mathematics (coupled with a glimmer of luck), I graduated with a Master's in Computer Vision and Robotics in 2016 when the now happening AI revolution started. Life has changed for the better ever since.
#machinelearning #deeplearning #aibites